Molecular structure determination from NMR spectroscopy

• US 8,779,770 B2
• Filed: 09/02/2009
• Issued: 07/15/2014
• Est. Priority Date: 09/04/2008
• Status: Active Grant

• Current Assignee: University of Bristol
• Original Assignee: University of Bristol
• Inventors: Butts, Craig P., Harvey, Jeremy
• Primary Examiner(s): ARANA, LOUIS M
• Application Number:
  US13/061,468
• Publication Number:
  US 20110210730A1
• Time in Patent Office: 1,777 Days
• Field of Search: 324/312, 324/310, 324/307, 324/309
• US Class Current: 324/312
• CPC Class Codes: G01N 24/087 Structure determination of a chemical compound, e.g. of a biomolecule such as a protein
• CPC Class Codes: G01R 33/4633 Sequences for multi-dimensional NMR
• CPC Class Codes: G01R 33/465 applied to biological material, e.g. in vitro testing

Abstract

Methods for the determination of the molecular structures of compounds are disclosed, the methods comprising obtaining NMR spectroscopic measurements of compounds, determining internuclear distances from the NMR measurements, and inputting the distances to an algorithm to determine probable structures. Optionally, constraints may be added and the algorithm repeated. Usually, the methods do not require comparisons to databases of spectra during the generation of possible structures.

Claims

1. A method for the determination of the molecular structure of a compound, the method comprising:

a) obtaining one or more nuclear magnetic resonance spectroscopic measurements from the compound, wherein the one or more nuclear magnetic resonance spectroscopic measurements are selected from;

X spectra, multidimensional X—

X correlation experiments, multidimensional X—

Y correlation experiments, nuclear Overhauser effect measurements, rotational Overhauser effect measurements, and a combination thereof;

wherein X and Y refer to an active NMR nucleus, and wherein at least one of;

the X—

X correlation experiments and the X—

Y correlation experiments, are selected from;

1-bond correlation experiments, 2-bond correlation experiments, multiple bond correlation experiments, and a combination thereof;

b) determining internuclear distances from the nuclear magnetic resonance spectroscopic measurements; and

c) inputting internuclear distances to a distance-geometry algorithm to determine the probable structure(s) of the compound, wherein the compound has a molecular weight of less than 2000 Daltons and is not a protein.

2. A method as claimed in claim 1, wherein X and Y are independently selected from ¹H, ¹³C, ¹⁵N, ¹⁹F, ³¹P, ¹¹B, ²⁹Si, ¹⁷O or other NMR active nuclei.

3. A method as claimed in claim 1, wherein the X spectra are 1 dimensional ¹H spectra, ¹³C spectra and/or ¹⁵N spectra.

4. A method as claimed in claim 1, wherein the X—

X and/or X—

Y correlation experiments are selected from one or more of Heteronuclear Single Quantum Coherence (HSQC), HSQC —

Total Correlation Spectroscopy (HSQC-TOCSY), Heteronuclear Multiple Bond Correlation (HMBC), Heteronuclear 2 Bond Correlation (H2BC), Incredible Natural Abundance Double Quantum Transfer Experiment (INADEQUATE), Correlated Spectroscopy (COSY) and Total Correlation Spectroscopy (TOCSY) experiments.

5. A method as claimed in claim 1, wherein the nuclear Overhauser effect or rotational Overhauser effect measurements are selected from one or more of NOE or ROE, and X—

X NOE or ROE measurements, wherein X is preferably ¹H.

6. A method as claimed in claim 1, further comprising a step of estimating internuclear distances by assuming average, standard bond distances based on connectivity from 1-bond X—

X or X—

Y correlation experiments and inputting the estimated distances to the distance-geometry algorithm.

7. A method as claimed in claim 1, further comprising a step of estimating typical bond or through-space distances based on connectivity from 2-bond correlation experiments and inputting the estimated distances to the distance-geometry algorithm.

8. A method as claimed in claim 1, further comprising a step of determining estimated ranges of values of distances based on multiple bond correlation experiments and inputting the estimated maximum values to the distance-geometry algorithm.

9. A method as claimed in claim 1, wherein determining the internuclear distances is from the NOE and/or ROE measurements.

10. A method as claimed in claim 1, further comprising the step, before obtaining the NMR spectroscopic measurements, of providing a compound and dissolving the compound in an NMR acceptable solvent to obtain a solution.

11. A method as claimed in claim 10, wherein the NMR acceptable solvent is a deuterated solvent.

12. A method as claimed in claim 1, wherein the compound contains C and H atoms.

13. A method as claimed in claim 1, wherein the distance-geometry algorithm is computer implemented.

14. A method as claimed in claim 1, wherein the method does not comprise a step of comparing the NMR spectroscopic measurements to a computer database of known NMR spectroscopic measurements, nor known molecular structures, during the generation of possible structural candidates.

15. A method as claimed in claim 1, which comprises an additional step (d) inputting a constraint to the algorithm and repeating step (c).

16. A method as claimed in claim 15, wherein inputting a constraint comprises at least one of:

(i) modifying distance constraints for H—

C and C—

C distances to account for the hybridisation at carbon centers;

(ii) constraining known chemical features to physically reasonable geometries;

(iii) introducing at least one of;

distances of other nuclei, and connectivity of other nuclei;

(iv) modifying Type 2 distance ranges to recognize the bonding which is apparent in the structure solution; and

(v) adding distances based on additional spectroscopic information available.

17. A method as claimed in claim 1, wherein the X—

X and/or X—

Y correlation experiments are selected from one or more of;

Heteronuclear Multiple Bond Correlation (HMBC), Heteronuclear 2 Bond Correlation (H2BC), Incredible Natural Abundance Double Quantum Transfer Experiment (INADEQUATE), and Correlated Spectroscopy (COSY) experiments.

18. A method as claimed in claim 1, wherein the distance-geometry algorithm comprises global optimization of a figure of merit function relating structure to the goodness of fit between distances in the structure and an input distance matrix.

19. A method as claimed in claim 18, wherein the merit function comprises a sum of distance constraint terms.